Proteins at Interfaces II - American Chemical Society

probe) and condensed phases appear dark (probe depleted regions). ... fluorescence image of a DPPC monolayer compressed beyond the phase transition...
0 downloads 0 Views 2MB Size
Chapter 35

Phospholipase A Interactions with Model Lipid Monolayer Membranes at the Air—Water Interface 2

1

2

2

3,5

K. M . Maloney , M . Grandbois , C. Salesse , D. W. Grainger , and A. Reichert 4

1

Downloaded by FUDAN UNIV on February 13, 2017 | http://pubs.acs.org Publication Date: May 5, 1995 | doi: 10.1021/bk-1995-0602.ch035

Department of Chemistry, Biochemistry, and Molecular Biology, Oregon Graduate Institute of Science and Technology, Portland, OR 97291-1000 Centre de Recherche en Photobiophysique, Université du Québec àTrois-Rivières, 3351 boulevard des Forges, C.P. 500, Trois-Rivières, Québec G9A 5H7, Canada Department of Chemistry, Colorado State University, Fort Collins, CO 80523 Materials Science Division, Lawrence Berkeley Laboratory, Berkeley, CA 94720 2

3

4

Mechanisms of domain formation by the membrane-active enzyme, phospholipase A2 (PLA ), in hydrolyzed phospholipid monolayers at the air-water interface have been investigated. PLA -catalyzed hydrolysis of phospholipid membrane substrates yields both free fatty acid and lyso-lipid reaction products. After a certain extent of monolayer hydrolysis, PLA forms large, regular two-dimensional aggregates at the monolayer interface. PLA -catalyzed monolayer hydrolysis has been investigated using monolayer isotherm techniques, surface potential, and dual-label fluorescence microscopy at the air-water interface, as well as electron microscopy. Our results indicate that PLA -produced fatty acid reaction products laterally phase separate from remaining enzyme substrate and lyso-lipid, forming two-dimensional, anionic domains. Results support our hypothesis that enzyme domains are formed by PLA binding electrostatically to phase separated fatty acid through basic amino acid residues located on the enzyme's interfacial binding surface. 2

2

2

2

2

2

Monolayers at the air-water interface (1) have been used as organized molecular scaffolds for two-dimensional protein crystallization (2), and as the basic structural unit for biosensing devices (5). As model membrane systems, monolayers have also been used to investigate specific protein-membrane interactions in natural and synthetic systems (4). Membrane lateral density and phase behavior are 5

Corresponding author 0097-6156/95/0602-0490$12.00/0 © 1995 American Chemical Society Horbett and Brash; Proteins at Interfaces II ACS Symposium Series; American Chemical Society: Washington, DC, 1995.

35. MALONEY ET AL.

PLA Interactions with Lipid Monolayer Membranes 2

491

experimentally accessible parameters, allowing the determination of macromolecule-membrane interactions as functions of monolayer membrane surface density and phase behavior. Application of surface characterization techniques such as scanning probe microscopy, quartz crystal microbalance, optical spectroscopy, and scattering methods, as well as traditional monolayer isothermal compression and fluorescence microscopy allow in-depth characterization of monolayer and monolayer-protein assemblies (see ref. 5 for a review). Naturally, a cascading series of biological events often results from protein membrane recognition, binding and subsequent biochemical events. Biological signal transduction from phospholipase A2 (PLA ) is one example (6). Calcium-dependent, secreted P L A ' s are small ( M « 14 kD), water-soluble enzymes derived from many sources, most common of which are mammalian pancreas and snake venoms. P L A catalyzes the stereospecific hydrolysis of sn-2 phospholipid acyl chains to produce a membrane-resident fatty acid and lyso-lipid species (7). Frequently, the fatty acid released is arachidonic acid, a precursor in the potent eicosanoid pathway (8). Extensive efforts have been made to elucidate the interfacial mechanism of P L A hydrolytic activity (9). It is well known that the enzyme displays maximal activity towards organized lipid interfaces such as vesicles, micelles, and monolayers (10). The influence of the physical state of the lipid interface on enzyme activity has been the focus of our investigations. We and coworkers have focused on elucidating the mechanism of P L A domain formation following lipid membrane hydrolysis at the air-water interface (11-14). We present here a mini-review of our work relating to P L A domain formation in these systems. We discuss our initial enzyme self-assembly results, an experimental system aimed at testing our domain formation hypothesis, and finally, our attempts to mimic P L A domain formation using a model mixed monolayer system with a water-soluble cationic dye protein analog. 2

Downloaded by FUDAN UNIV on February 13, 2017 | http://pubs.acs.org Publication Date: May 5, 1995 | doi: 10.1021/bk-1995-0602.ch035

2

r

2

2

2

2

2

Experimental Materials. Rhodamine-labeled phosphatidylethanolamine or l-palmitoyl-2[12-(7-nitro-2-l,3-benzoxadiazol-4-yl)amino]dodecanoyl-phosphatidylcholine (Rhod-DPPE and C12-NBD-PC, respectively) and cationic dye 1,1',3,3,3',3'hexamethylindocarbo-cyanine iodide (H-379), were purchased from Molecular Probes (Eugene, OR) and used as received. Nonfluorescent lipids and fatty acids were from Avanti Polar Lipids (Alabaster, A L ) and Fluka A G , respectively. The asymmetric lipids 6,16-PC and 16,6-PC used in the hydrolysis experiments were synthesized via the fatty acid imidazole method (15,16), and were generously synthesized, purified, and supplied by Prof. Mary Roberts (Boston College). Naja N.N P L A was purchased from Sigma and used as received. 2

Enzyme Labeling. PLA -fluorescein (PLA -FITC) conjugates were prepared by the general method of Nargessi and Smith (17). Labeling efficiencies were statistical, with [FITC]/[PLA ] ratios typically less than 0.75. FITC concentrations were determined by U V absorbance at 490 nm and protein concentration determined by the method of Smith and coworkers (18). 2

2

2

Horbett and Brash; Proteins at Interfaces II ACS Symposium Series; American Chemical Society: Washington, DC, 1995.

PROTEINS AT INTERFACES II

492

Fluorescence Microscopy. Fluorescence microscopy of monolayers at the air-water interface was conducted on home-built, microprocessor-controlled Teflon troughs mounted on a stage of a Zeiss A C M epifluorescence microscope (19). Image contrast of lipid monolayers by fluorescence microscopy is based on differential solubilities of fluorescentiy-labeled lipid probes between monolayer phases (20-22). In phase-separated monolayer systems exhibiting a coexistence of fluid and condensed phases, Rhod-PE and C12-NBD-PC, preferentially partition into fluid monolayer phases. Thus, fluid phases appear bright (enriched with fluorescent lipid probe) and condensed phases appear dark (probe depleted regions). Monolayer probe concentrations were between 0.5 and 1 mol%. Using excitation and emission specific fluorescence filters integrated with the fluorescence microscope, it is possible to conduct dual-label imaging experiments with rhodamine-labeled lipid and fluorescein-labeled P L A (77,72). Specific details concerning fluorescence microscopy of lipid monolayers at the air-water interface are found elsewhere

Downloaded by FUDAN UNIV on February 13, 2017 | http://pubs.acs.org Publication Date: May 5, 1995 | doi: 10.1021/bk-1995-0602.ch035

2

(19,22). Dye Binding to Monolayers. The water soluble, cationic dye H-379 was used to probe electrostatic characteristics of phase separated monolayers. H-379 was injected beneath the monolayer and into a Teflon mask positioned in the trough (11,14). The mask facilitated monolayer imaging by reducing monolayer surface flow. Monolayer images from inside and outside the mask were frequently compared to ensure the mask did not create any monolayer artifacts. Lipid monolayer imaging was achieved with C12-NBD-PC probe and was directly visualized with a fluorescein specific fluorescence filter. Switching between fluorescein and rhodamine specific filters allowed tandem imaging of lipid monolayer physical state and H-379 dye location, respectively. Surface Potential. Monolayer surface potential was determined using the ionizing electrode method (7). An A m electrode was placed 1-2 mm above a monolayer covered interface. A Pt electrode in the monolayer subphase was used as a reference electrode. Surface potential and molecular area at constant surface pressure (determined with a Wilhelmy plate) were measured simultaneously as a function of time. Surface potential, A V (mV), is defined as the difference in potential of a monolayer film versus a clean interface. During hydrolysis experiments, the monolayer subphase was stirred. 2 4 1

Electron Microscopy. L-a-DPPC monolayers were hydrolyzed by P L A at 22 mN/m on a Tris-buffered subphase (30 °C). PLA -hydrolyzed DPPC monolayers were transferred to carbon-coated E M grids by the Langmuir-Schaefer technique (23). Protein staining was achieved using a 2% uranyl acetate solution for 30 seconds. 2

2

Results Dipalmitoylphosphatidylcholine (DPPC) monolayers at the air-liquid interface exhibit several phase transitions while undergoing isothermal compression. While great debate has taken place over whether or not any of the observed transitions are truly first-order (4), fluorescence and Brewster angle microscopy at the air-liquid interface have unequivocally shown a coexistence of liquid-expanded and solid

Horbett and Brash; Proteins at Interfaces II ACS Symposium Series; American Chemical Society: Washington, DC, 1995.

35.

M A L O N E Y ET A L .

PLA Interactions with Lipid Monolayer Membranes 2

493

monolayer phases in the near-horizontal region of the DPPC isotherm. A fluorescence image of a DPPC monolayer compressed beyond the phase transition point into this phase coexistence region (19-22) is shown in Figure la, as viewed through a rhodamine-specific (monolayer) fluorescence filter (ll 12). Clearly visible are dark regions depleted of monolayer fluorescent probe solid domains surrounded by fluid L E lipid phase (fluorescent probe rich). FITC-labeled P L A was then injected into the DPPC monolayer subphase beneath the monolayer (77). As viewed through a fluorescein-specific fluorescence filter, Figure lb shows a nearly homogeneous fluorescence signal from F I T C - P L A at the monolayer interface. As time after P L A injection increases, solid domains are hydrolyzed by the enzyme, as evidenced in Figures lc and e. The interior of solid DPPC domains appear to be slowly degraded as the enzyme hydrolyzes at the interface between the solid DPPC domains*and L E phase DPPC. However, as shown in Figure Id (25 minutes after P L A injection), the F I T C - P L A fluorescence signal is no longer homogeneous. Small areas of concentrated fluorescence are visible in the monolayer plane. These small areas represent aggregated F I T C - P L A at the hydrolyzed monolayer interface. As hydrolysis time increases, the enzyme domains increase in size (Figure If, 40 minutes after P L A injection) and number. Moreover, the fluorescent P L A domains in Figure If are also seen when the lipid monolayer is imaged with a rhodamine filter (non-fluorescent gray regions in Figure le corresponding to bright regions in Figure If). Finally, 60 minutes after P L A introduction to the monolayer subphase, fluorescent P L A domains are quite large (Figure lh) with the fluorescence signal corresponding exactly to the dark bean-shaped regions in Figure lg. P L A domains depicted in Figure 1 resist lateral compression and dissolution at high and low surface pressures, respectively (12). This is consistent with tight enzyme packing and lateral enzyme association in the interfacial plane. f

2

2

Downloaded by FUDAN UNIV on February 13, 2017 | http://pubs.acs.org Publication Date: May 5, 1995 | doi: 10.1021/bk-1995-0602.ch035

2

2

2

2

2

2

2

2

2

In an attempt to mimic P L A domain formation, we characterized binary and ternary mixed monolayers containing DPPC, C16Lyso and palmitic acid (PA). Ternary mixed monolayers represent the composition of P L A hydrolyzed DPPC monolayers at various extents of hydrolysis. Initially, the mole fraction of DPPC is unity. As a function of increasing P L A hydrolysis time, mole fraction of DPPC approaches zero, while mole fractions of C16Lyso and PA (enzymatically produced in equimolar amounts) each approach 0.5. Representing various extents of hydrolysis, mixed monolayers of C16Lyso:PA (1:1; mokmol) and DPPC:C16Lyso:PA (0.2:1:1, 1:1:1, 2:1:1, up to 10:1:1; mol:mol:mol) were investigated (14). C16Lyso:PA (1:1) binary mixed monolayers represent a totally hydrolyzed lipid monolayer while DPPC:C16Lyso:PA ternary mixed monolayers represent partially hydrolyzed monolayer compositions. Shown in Figure 2a is a fluorescence micrograph of ternary mixed DPPC:C16Lyso:PA (0.2:1:1) monolayers (no enzyme). Compression of the monolayer yields the phase separated microstructure shown in Figure 2. Injection of cationic, water-soluble H-379 dye results in rapid dye binding to the phase separated microstructure (Figure 2b), showing the domain region is enriched in anionic components (fatty acid). As mole fraction of DPPC increases, the size of phase separated anionic domains decreases, and is absent on the micrometer scale once a 3:1:1 (DPPC:C16Lyso:PA) ternary mixed monolayer ratio is obtained. Phase separation of anionic domains such as that shown in Figures 2a and b are only observed in the presence of Ca * and on an alkaline monolayer subphase. This is consistent with a Ca -mediated fatty acid 2

2

2

2

2+

Horbett and Brash; Proteins at Interfaces II ACS Symposium Series; American Chemical Society: Washington, DC, 1995.

PROTEINS AT INTERFACES II

Downloaded by FUDAN UNIV on February 13, 2017 | http://pubs.acs.org Publication Date: May 5, 1995 | doi: 10.1021/bk-1995-0602.ch035

494

Figure 1. Fluorescence micrographs of P L A hydrolyzed DPPC monolayers. Figures a, c, e, and g were imaged through a filter specific for rhodamine fluorescence (monolayer lipid probe). Figures b, d, f, and h were imaged through a fluorescein specific fluorescence filter. Image pairs a-b, c-d, e-f, and g-h were taken 0, 25, 40, and 60 minutes after P L A injection into the monolayer subphase, respectively. Subphase is 100 mM NaCl, 10 m M Tris, 5 m M C a C l , pH 8.9, T = 3 0 ° C . Surface pressure (22 mN/m) remained constant throughout hydrolysis. Scale bar in a is 20 /xm. Reprinted with permission from Elsevier Scientific Press. 2

2

2

Horbett and Brash; Proteins at Interfaces II ACS Symposium Series; American Chemical Society: Washington, DC, 1995.

MALONEY ET AL.

PLA Interactions with Lipid Monolayer Membranes 2

Downloaded by FUDAN UNIV on February 13, 2017 | http://pubs.acs.org Publication Date: May 5, 1995 | doi: 10.1021/bk-1995-0602.ch035

35.

Figure 2. Fluorescence micrographs of phase separated, ternary mixed monolayers (DPPC:C16Lyso:PA, 0.2:1:1). Images in a and b as viewed through a fluorescein (monolayer filter) and rhodamine (H-379) specific fluorescence filter, respectively, at 12 mN/m. Monolayer subphase is 100 mM NaCl, 10 mM Tris, 5 mM C a C l , pH 8.9, T = 2 0 ° C . Scale bar = 25 ^m. 2

Horbett and Brash; Proteins at Interfaces II ACS Symposium Series; American Chemical Society: Washington, DC, 1995.

495

PROTEINS AT INTERFACES II

496

chelating event. Alkaline subphase pH is required to keep fatty acid carboxylate head-groups ionized (pK ca. 5-6). By contrast, replacing palmitic acid in ternary mixed monolayer systems with its unsaturated analog, palmitoleic acid, does not yield phase separation of fatty acid enriched microstructures upon monolayer compression (unpublished data). Palmitoleic acid is unable to pack into condensed phases due to its acyl chain unsaturation. Additionally, phase separated polymerized diacetylenic fatty acid (DIAC-FA) surrounded by fluid D-DPPC lipid matrix does not lead to P L A adsorption (75). However, cationic dye adsorption does occur under these phase separated D I A C - F A monolayer domains, but presumably during D I A C - F A phase separation and polymerization other monolayer species important for P L A binding (i.e., PC) are expelled from phase separated D I A C - F A (75). This shows that fatty acid chemistry and the presence of substrate P C both influence protein interfacial domain formation. a

2

Downloaded by FUDAN UNIV on February 13, 2017 | http://pubs.acs.org Publication Date: May 5, 1995 | doi: 10.1021/bk-1995-0602.ch035

2

To account for P L A self-assembly at hydrolyzed phospholipid monolayers at the air-liquid interface, we and coworkers proposed that PLA -produced fatty acid reaction products were responsible for inducing interfacial enzyme domain formation (77,72). To test this hypothesis, the importance of fatty acids, as well as lyso-lipid and lipid substrate, on P L A domain formation was investigated using asymmetric phospholipid substrates. Hydrolysis of these substrates results in formation of selectively water-soluble fatty acids and water-insoluble lyso-lipids (Maloney, K . M . , Grandbois, M . , Grainger, D . W . , Salesse, C , Lewis, K . A . , and Roberts, M . F . , submitted to Biochimica et Biophysica Acta). Results of P L A hydrolysis of l-palmitoyl,2-caproyl-phosphatidylcholine (16,6-PC) monolayers are shown in Figure 3a. P L A hydrolysis of 16,6-PC produces water-soluble caproic acid and water-insoluble palmitoyl-lysophosphatidylcholine (C16Lyso). Total hydrolysis of a 16,6-PC monolayer by P L A thus results in formation of a pure C16Lyso monolayer. Plotted in Figure 3a is monolayer molecular area and surface potential as a function of P L A hydrolysis time. The horizontal regions of the surface potential and monolayer area curves (segments A B and A ' B ' , respectively) reflect the stability of 16,6-PC monolayers in the absence of enzyme. At point B, P L A was injected into the monolayer subphase. Immediately following P L A injection, surface potential and monolayer molecular area start to decrease. This decrease is linear as a function of time and continues to point C . At point C , surface potential and monolayer area continue to decrease, but at a slower rate. Point C is taken as the end of 16,6-PC monolayer hydrolysis, and the remaining slow decrease in surface potential and area reflecting the instability of the remaining C16Lyso monolayer. At the end of hydrolysis (point C), monolayer molecular area is 49.5 A /molecule and the surface potential is 248 mV, corresponding closely to molecular area and surface potential expected for a pure C16Lyso monolayer under these conditions (50 A /molecule and 220 mV, respectively). 2

2

2

2

2

2

2

2

2

2

2

Shown in Figure 3b is surface potential and molecular area versus P L A hydrolysis time of l-caproyl,2-palmitoyl-phosphatidylcholine (6,16-PC), the positional isomer of 16,6-PC. P L A hydrolysis of 6,16-PC produces water-soluble caproyl-lysophosphatidylcholine and water-insoluble palmitic acid. P L A hydrolysis of 6,16-PC monolayers results in the formation of a pure palmitic acid monolayer. Similar to Figure 3a, surface potential and molecular area are stable prior to enzyme introduction (segments A B and A ' B ' , respectively). After P L A injection beneath 2

2

2

2

Horbett and Brash; Proteins at Interfaces II ACS Symposium Series; American Chemical Society: Washington, DC, 1995.

MALONEY ET AL.

PLA Interactions with Lipid Monolayer Membranes 497 2

Downloaded by FUDAN UNIV on February 13, 2017 | http://pubs.acs.org Publication Date: May 5, 1995 | doi: 10.1021/bk-1995-0602.ch035

35.

Figure 3. Surface potential (—) and monolayer molecular area ( ) versus P L A hydrolysis time at constant surface pressure (15 mN/m). a and b represent 16,6-PC and 6,16-PC monolayers respectively. Monolayer subphase is 100 mM NaCl, 10 mM Tris, 5 m M CaCl , pH 8.9, T = 2 0 ° C , compression rate = 2 A /molecule/minute. P L A injected at point B. 2

2

2

2

Horbett and Brash; Proteins at Interfaces II ACS Symposium Series; American Chemical Society: Washington, DC, 1995.

PROTEINS AT INTERFACES II

498

the monolayer (point B), surface potential and monolayer area immediately decrease until point C (taken as end of hydrolysis) is reached. The molecular area at point C is 19 A /molecule and the surface potential is 100 mV. The area closely corresponds to the molecular area for a fatty acid molecule under these conditions, (ca. 20 A /molecule) while the surface potential value does not correspond to the value expected for a palmitic acid monolayer (-30 mV) under these conditions. As shown in Figures 4a and b, P L A was observed to form domains in hydrolyzed 6,16-PC monolayers. Figure 4a shows a gray, phase separated microstructure in a rhodamine specific fluorescence filter. Imaging the same field of view with a fluorescein filter shows that F I T C - P L A fluorescence corresponds exactly to the gray region in Figure 4a. By contrast, P L A hydrolysis of 16,6-PC monolayers results in the formation of a C16Lyso monolayer (fatty acid products are solubilized in the subphase) with little PLA domain formation. Moreover, addition of fatty-acid binding albumin to the monolayer subphase during P L A hydrolysis of 16,6-PC monolayers totally suppressed P L A interfacial aggregation. Surface potential and molecular area hydrolysis kinetics (Figure 3) show that P L A does not contribute to the monolayer molecular area. Enzyme domains shown in Figure 4 are located beneath the monolayer and remain stable for hours. Similarly, P L A domains shown in Figure 1 also lie directly beneath the hydrolyzed monolayer. Control experiments consisting of P L A injection beneath both pure palmitic acid (unpublished data) and palmitoyl-lysophosphatidyl-choline (72) monolayers were also conducted. P L A does not form domains beneath pure lyso-lipid monolayers. P L A does, however, form domains beneath pure palmitic acid monolayers (Ca -containing pH 8.9 subphase), though exhibiting highly irregular domain morphologies distinct from those formed in the presence of PC substrate. To further assess the microstructure of P L A domains, hydrolyzed monolayers were transferred to E M grids. Figure 5 shows an electron micrograph of a PLA -hydrolyzed L-DPPC monolayer (24). Small regions showing hexagonally ordered protein are visible. Experimental difficulties (transfer of fluid phase monolayers) prevented full transfer of intact P L A domains (ca. 25 fim in diameter). Hexagonally arranged single P L A crystals in Figure 5 most likely represent fragments of 2-D larger P L A domains. 2

2

2

2

2

Downloaded by FUDAN UNIV on February 13, 2017 | http://pubs.acs.org Publication Date: May 5, 1995 | doi: 10.1021/bk-1995-0602.ch035

2

2

2

2

2

2

2

2

2+

2

2

2

2

2

Discussion Figure 6 depicts our P L A domain formation hypothesis (adapted from ref. 11). P L A catalyzes the chemical conversion of a single component lipid monolayer to a ternary mixed, heterogeneous interface. We propose that after a certain extent of enzyme hydrolysis, fatty acid reaction products phase separate from remaining substrate and lyso-lipid. The result is the formation of anionic fatty acid enriched microstructures within the monolayer which prompt P L A interfacial binding. Figure 1 shows, as a function of hydrolysis time, the formation of these P L A microstructures. Previous accounts concerning analogous membrane microstructuring from P L A action on vesicle bilayers have been reported by Jain and coworkers (25,26). They showed ternary mixed vesicle systems undergo lateral phase segregation yielding bilayer regions enriched in fatty acids as monitored using 2

2

2

2

2

Horbett and Brash; Proteins at Interfaces II ACS Symposium Series; American Chemical Society: Washington, DC, 1995.

MALONEY ET AL.

PLA

2

Interactions with Lipid Monolayer Membranes

Downloaded by FUDAN UNIV on February 13, 2017 | http://pubs.acs.org Publication Date: May 5, 1995 | doi: 10.1021/bk-1995-0602.ch035

35.

Figure 4. Fluorescence images of P L A hydrolyzed 6,16-PC monolayers at 15 mN/m. Images in a and b are viewed through rhodamine and fluorescein filters, respectively. Monolayer subphase is 100 mM NaCl, 10 m M Tris, 5 mM C a C l , 25 n M BSA pH 8.9, T = 2 0 ° C . Scale bar in a is 25 p.m. 2

2

Horbett and Brash; Proteins at Interfaces II ACS Symposium Series; American Chemical Society: Washington, DC, 1995.

499

PROTEINS AT INTERFACES II

Downloaded by FUDAN UNIV on February 13, 2017 | http://pubs.acs.org Publication Date: May 5, 1995 | doi: 10.1021/bk-1995-0602.ch035

500

Figure 5. Electron micrograph of PLA -hydrolyzed L-DPPC monolayers. Hydrolyzed monolayers were transferred to E M grids via the Langmuir-Schaefer technique at 22 mN/m. Scale bar is 40 nm. 2

mm

D i

y Monolayer Hydrolysis

Hydrolyzed monolayer of lipid, lyso-lipid and fatty acid A ^

Adsorption

Figure 6. Protein domain formation hypothesis for PLA -hydrolyzed phospholipid monolayers at the air-water interface. See text for details. 2

Horbett and Brash; Proteins at Interfaces II ACS Symposium Series; American Chemical Society: Washington, DC, 1995.

35. MALONEY ET AL.

PLA Interactions with Lipid Monolayer Membranes 2

501

cationic dye binding and fluorescence techniques and P L A hydrolysis (25,26). Though P L A domain formation has not been shown in vesicle systems, acceleration of enzyme hydrolysis kinetics has been correlated with vesicle fatty acid lateral segregation (26). Both monolayers and vesicles share similar physical membrane characteristics after limited PLA2 hydrolysis: laterally phase separated membrane regions enriched in fatty acid. More recently, further evidence supporting lateral phase separation in ternary vesicle systems has also been presented by Biltonen and coworkers using fluorescence techniques (27). To test the enzyme domain formation hypothesis in monolayers, we employed asymmetric lipid substrates: enzymatic hydrolysis of 6,16-PC substrate leads to the formation of water-soluble lyso-lipid and water-insoluble fatty acids. P L A domain formation in this system was observed, indicating that lyso-lipid does not play a critical role in P L A interfacial aggregation. Hydrolysis of 16,6-PC, on the other hand, leads to the formation of a water insoluble lyso-lipid, and a water-soluble fatty acid. P L A domain formation in this system was greatly reduced, consistent with our hypothesis that fatty acids are necessary for enzyme aggregation. In addition, the time needed to observe P L A domain in 16,6-PC monolayer systems was typically 8-10 hours after P L A injection, whereas P L A domain formation in hydrolyzed 6,16-PC systems was usually less than one hour. Since small P L A domains were observed in hydrolyzed 16,6-PC systems, bovine serum albumin (BSA) was added to the monolayer subphase during hydrolysis to sequester fatty acid. With BSA present in the subphase, P L A domain formation was completely suppressed. Though all caproic acid reaction products were probably solubilized after their hydrolytic release, aqueous self-acylation of P L A is known to occur (28). Perhaps the long induction times needed for P L A domain formation with the 16,6-PC substrate result from this acylation mechanism (and resulting increase in surface activity of P L A ^ . Surface potential and molecular area versus hydrolysis time (Figure 3) also support our claim that fatty acids and lyso-lipids are solubilized upon P L A hydrolysis for 16,6-PC and 6,16-PC, respectively. If lyso-lipid were important for P L A domain formation, P L A domain formation would have either been reduced or totally suppressed in hydrolyzed 6,16-PC systems. As shown in Figure 4, this was not the case. Recently solved high-resolution P L A crystal (29-31) structures show that several cationic amino acid residues (Lys, Arg) located near the enzyme's interfacial recognition site are solvent accessible. It is reasonable to expect that these residues possess the ability to interact with negative charges concentrated within phase separated fatty acid domains. In fact, P L A from porcine pancreas is well-known to prefer substrate present in a negatively charged interface (9). Additional support for electrostatic interactions between P L A and fatty acids is given by surface potential data for P L A hydrolyzed 6,16-PC monolayers. The final surface potential obtained after completion of 6,16-PC hydrolysis was 100 mV, which does not correspond to that of a pure palmitic acid monolayer under enzyme-free conditions (-30 mV). As shown in Figure 4, P L A domain formation occurred during 6,16PC hydrolysis. Therefore, P L A ' s dipoles and surface charges also contribute to the surface potential. Though we cannot separate P L A ' s and palmitic acid's contribution to the surface potential, this suggests P L A electrostatically interacts 2

2

2

Downloaded by FUDAN UNIV on February 13, 2017 | http://pubs.acs.org Publication Date: May 5, 1995 | doi: 10.1021/bk-1995-0602.ch035

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

Horbett and Brash; Proteins at Interfaces II ACS Symposium Series; American Chemical Society: Washington, DC, 1995.

PROTEINS AT INTERFACES II

502

with fatty acids. Results from ternary mixed monolayer studies also show that fatty acids play an important role in the phase behavior of resultant P L A hydrolyzed lipid monolayers. Using H-379 as a cationic protein "analog," Figure 2 shows that under appropriate conditions (high pH, presence of C a ) , ternary mixed monolayers representing lipid membrane compositions at various extents of hydrolysis undergo fatty acid phase separation (13,14). Portions of P L A domains transferred to electron microscopy grids apparently exhibit 2-D hexagonal order. Collectively, these observations support our hypothesis that P L A domain formation critically depends on the presence of fatty acids at the air-water interface. The shapes of P L A domains in hydrolyzed DPPC and 6,16-PC films (Figures 1 and 4, respectively) are dissimilar. It is therefore possible that two different P L A aggregation mechanisms are occurring in hydrolyzed DPPC and 6,16-PC systems. As shown in Figure 1, P L A domains from hydrolyzed isomerically pure lipid monolayers are typically bean-shaped. Similar bean-shaped domains are also observed in PLA -Jree, phase separated ternary mixed monolayers containing enantiomerically pure DPPC, C16Lyso and palmitic acid (0.2:1:1, Figure 2). The well-characterized property of DPPC enantiomers to induce chiral lipid domains during lipid monolayer phase transitions (19-22) may also play a role in templating or structuring P L A interfacial aggregation into specific morphologies. This property would also extend to DPPC's chiral hydrolysis product, C16Lyso lipid, resident after P L A hydrolysis even though it lacks an observable phase transition and does not form monolayer domains (14). Nevertheless, hydrolysis of asymmetric phospholipid substrates (Figure 4) results in P L A domains that are irregularly shaped and often comprise multiple domains clustered together. To obtain the regular bean-shaped P L A domains after hydrolysis, C16Lyso and/or DPPC residual in phase separated fatty acid regions may play an important role in monolayer phase separation, enzyme domain formation and its resulting morphology. In 6,16-PC hydrolyzed monolayers, however, lyso-lipid products are solubilized during hydrolysis; the remaining monolayer is pure fatty acid. In this case, P L A domain formation is still observed but with distinctly different morphology (Figure 4). It is our assertion in this case that P L A adsorbs interfacially to fatty acid without the morphology imposed by chiral lipids. This is supported by control experiments showing P L A adsorbs to pure palmitic acid monolayers resulting in highly irregular protein domain morphologies. We therefore alter our original enzyme domain formation hypothesis to include chiral PC in phase separated monolayer regions and its influence on resulting domani morphologies. P L A adsorbs to pure palmitic acid monolayers, but to observe chiral protein domains (12), enzyme substrate is necessary at some point during the enzyme domain formation time course. Further work is aimed at elucidating these points. 2

2 +

2

2

2

2

Downloaded by FUDAN UNIV on February 13, 2017 | http://pubs.acs.org Publication Date: May 5, 1995 | doi: 10.1021/bk-1995-0602.ch035

2

2

2

2

2

2

2

2

2

2

Acknowledgments We acknowledge the inspiration and creativity of Prof. Helmut Ringsdorf (University of Mainz) and his continued support of our work. We also thank Prof. Mary Roberts (Boston College) for synthesizing, purifying, and supplying the asymmetric phospholipids. Financial support from National Science Foundation

Horbett and Brash; Proteins at Interfaces II ACS Symposium Series; American Chemical Society: Washington, DC, 1995.

35. MALONEY ET AL.

PLA Interactions with Lipid Monolayer Membranes 2

503

grant DMR-9357439 (DWG) and grant INT-9303588 (KMM), Natural Sciences and Engineering Research Council of Canada, The Fonds F C A R , and the FRSQ (CS), a N A T O Travel Grant (DWG and CS), and scholarships from the Fonds F C A R (MG) is greatly appreciated.

Literature Cited 1.

Downloaded by FUDAN UNIV on February 13, 2017 | http://pubs.acs.org Publication Date: May 5, 1995 | doi: 10.1021/bk-1995-0602.ch035

2. 3. 4. 5.

6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18.

19. 20. 21. 22. 23. 24.

Gaines, G. L., Jr. In Insoluble Monolayers at Liquid-Gas Interfaces; Prigogine, I., Ed.; John Wiley & Sons, Inc.: New York, NY, 1966; pp 73-79, 83-89. Uzgiris, E. E. and Kornberg, R. D. Nature 1983, 301, 125-129. Okahata, Y., Tsuruta, T., Ijiro, K., and Arigu, K. Thin Solid Films 1989, 180, 65-72. Mohwald, H. Ann. Rev. Phys. Chem. 1990, 41, 441-476. Ulman, A. In An Introduction to Ultrathin Films: From Langmuir-Blodgett to Self-Assembly; Harcourt, Brace, Jovanovich: Boston, MA, 1991; pp 1-83. Mato, J. M. In Phospholipid Metabolism in Cellular Signaling; Mato, J. M., Ed.; CRC Press: Boca Raton, FL, 1990, pp 43-60. Waite, M. In Handbook of Lipid Research, Hanahan, D. J., Ed.; Plenum Press: New York, NY, 1987, Vol. 5; pp 155-241. Smith, W. L. Biochem. J. 1989, 259, 315-324. Jain, M. K. and Berg, O. G. Biochim. Biophys. Acta. 1989, 1002, 127-156. Pieterson, W. A.; Vidal, J. C.; Volwerk, J. J.; deHaas, G. H. Biochemistry 1974, 13, 1455-1460. Grainger, D. W.; Reichert, A.; Ringsdorf, H.; Salesse, C. Febs. Lett. 1989, 252, 79-85. Grainger, D. W.; Reichert, A.; Ringsdorf, H.; Salesse, C.Biochim. Biophys. Acta. 1990, 1023, 365-379. Reichert, A.; Wagenknecht, A.; Ringsdorf, H. Biochim. Biophys. Acta. 1992, 1106, 178-188. Maloney, K. M.; Grainger, D. W. Chem. Phys. Lipids. 1993, 65, 31-42. Burns, R. A.; Roberts, M. F. Biochemistry 1980, 19, 3100-3106. Burns, R. A.; Donovan, J. H.; Roberts, M. F. Biochemistry 1983, 11, 964-973. Nargessi, R. D.; Smith, D. S. Method. Enzymol. 1986, 122, 67-72. Smith, P. K.; Krohn, R. I.; Hermanson, G. T.; Mallia, A. K.; Gartner, F. H.; Provenzano, M. D.; Fujimoto, E. K.; Goeke, N. M.; Olson, B. J.; Klenk, D. C. Anal. Biochem. 1985, 150, 76-85. Meller, P. Rev. Sci. Instrum. 1988, 59, 2225-2231. McConnell, H. M.; Weiss, R. M.; Tamm, L. K. Proc. Natl. Acad. Sci. USA 1984, 81, 3249-3253. Peters, R.; Beck, K. Proc. Natl. Acad. Sci. USA 1983, 81, 3249-3253. Losche, M.; Sackmann, E.; Mohwald, H. Ber. Bunsenges. Phys. Chem. 1983, 87, 848-852. Langmuir, I.; Schaefer, V. J.; Wrinch, D. M. Science 1937, 85, 76-80. Reichert, A. Ph.D. Dissertation, University of Mainz, Germany, 1993.

Horbett and Brash; Proteins at Interfaces II ACS Symposium Series; American Chemical Society: Washington, DC, 1995.

PROTEINS AT INTERFACES II

504

25. 26. 27. 28. 29. 30.

Downloaded by FUDAN UNIV on February 13, 2017 | http://pubs.acs.org Publication Date: May 5, 1995 | doi: 10.1021/bk-1995-0602.ch035

31.

Yu, B.-Z.; and Jain, M. K. Biochim. Biophys. Acta 1989, 980, 15-22. Jain, M.K.; Yu, B.-Z.; Kozubek, A. Biochim. Biophys. Acta 1989, 980, 23-32. Burack, W. R.; Yuan, Q.; and Biltonen, R. L. Biochemistry 1993, 32, 583-589. Cho, W.; Tomasselli, A. G.; Heinrikson, R. L.; Kezdy, F. J. J. Biol. Chem. 1988, 263(23), 11237-11241. White, S. P.; Scott, D. L.; Otwinowski, Z.; Gelb, M. H.; Sigler, P. B. Science, 1990, 250, 1560-1563. Scott, D. L.; Otwinowski, Z.; Gelb, M. H.; and Sigler, P. B. Science 1990, 250, 1563-1566. Dijkstra, B. W.; Kolk, K. H.; Hol, W. G. J.; Drenth, J. J. Mol. Biol. 1981, 147, 97-123.

RECEIVED December 22, 1994

Horbett and Brash; Proteins at Interfaces II ACS Symposium Series; American Chemical Society: Washington, DC, 1995.